A description of different methods for computer simulations of fast moving objects is given by Weiskopf . Here we will mention only one of them, spacetime ray-tracing, a method important for its universal applicability.
In a ray-tracing computation the image is divided into pixels. For each pixel, one determines the direction of the incoming light ray that reaches this pixel after crossing the lens. This light ray is then retraced from the camera into the scene. If the light ray hits an object, its defining pixel is given the colour of the object at the point of impact. In "normal" computer graphics applications, the light travel times can be neglected with respect to the motion of the objects and the camera. It is therefore adequate to search the points of intersection between the light rays and static objects in three-dimensional space. For relativistic simulations the ray-tracing method is extended to four-dimensional spacetime by retracing each light ray in space and time searching for events of intersection, i. e. for positions and times of encounters between light rays and objects.
For the computation of colour and brightness of objects the emitted spectrum (more precisely: the specific intensity) must be given at every point on the object surface. For each pixel on the image plane, the incident spectrum is transformed into the rest frame of the camera and is then converted into a colour impression by using the empirically determined laws of colorimetry.
Contact: Would you like to be notified of new contributions or to send us a message?
Survey: How do you use this site and how would you like it to develop?